DESCRIPTION (taken from the application) H. pylori strains are highly diverse, and humans can be infected with one or more individual strains. This project will examine horizontal gene transfer among strains in vitro and then in vivo. The work proposed stems from the observations that H. pylori are naturally competent for transformation, and the recent identification of conjugation in the investigator's laboratory. These data provide mechanisms that can help explain the diversity of strains, that can be used to develop classification systems for H. pylori, and that can explain the adaptation of populations of organisms to the changing environments in human stomachs during decades of colonization. The experiments planned will assess wild type and specific H. pylori mutants for their ability to recombine and to characterize the compatibility of strains and the mechanisms involved in recombination. Studies then are planned in rodents that can be chronically (months) colonized with H. pylori to examine the relationship between recombination and phenotypic changes in vivo. The animal models provide an approach to considering adaptation of these organisms to hosts of differing genotypes, by performing experiments with pairs of defined inbred mouse strains. Examination of expression of Lewis antigens by the bacteria in several hosts (including mice transgenic for Le-b expression by gastric epithelial cells) will permit exploration of the role of host genotype on bacterial phenotypic expression. Infecting animals with pairs of strains, then altering host conditions (gastric acidity or immune response), and observing shifts in the equilibrium between the pairs, and rates of observing new recombinants, provides an experimental format to address questions fundamental to the population biology and pathogenesis of H. pylori infections.